Our work began with our findings that the forkhead tanscription factor 2 (Foxl2) is important in preventing POF; most strikingly, we showed that Foxl2 hereditary deficiency provokes POF in women with the Blephariphomisis-Ptosis-Epicanthus Inversus Syndrome (BPES). We have then shown over the last 6 years that Foxl2 is critical throughout female reproductive life, vitally involved in the ovary in somatic sex determination and maintenance as well as in the development and stabilization of ovarian follicles. The studies have led to a significant change in the paradigm for somatic sex determination and maintenance, from a formulation in which the ovary was a default pathway when the male-determining gene Sry was absent or inactive to a model in which Sry remains male-determinative but several other genes, and especially Foxl2, are actively required to determine ovary formation and stability, continually repressing the male pathway. We have continued to analyze the action of Foxl2 by generating mouse models in which Foxl2 is ablated or overexpressed, either alone or in combination with other important ovarian factors. We have now progressed further in the elucidation of the role of Foxl2 in the formation and maintenance of follicles and in germ cell survival, and are extending the studies to the comparative action of other master genes affecting follicle dynamics and the results of joint ablation of several of them. In an extension of studies, we demonstrated that Cyp26b1, as well as transcription factor genes steroidogenic factor-1 (Sf1) and Sox9 are co-expressed in Sertoli cells in mouse fetal testes (and Cyp26b1 and Sf1 are co-expressed in Leydig cells). Furthermore, qRT-PCR showed that Cyp26b1 up-regulation by Sox9/Sf1 was attenuated by Foxl2, whereas in Foxl2-null mice, Cyp26b1 expression in XX gonads was increased 20-fold relative to WT controls. These data support the hypothesis that Sox9 and Sf1 ensure the male fate of germ cells in part by up-regulating Cyp26b1, and that Foxl2 antagonizes Cyp26b1 expression (and Sox9) in ovaries. We also investigated the role of bone morphogenetic protein 2 (Bmp2) and Foxl2 in the regulation of a target gene, follistation (Fst) expression in the mouse ovary. We found that Bmp2 and Foxl2 cooperate to promote correct expression throughout the developmental program of the ovary. Moreover, study of Wnt4 knock-out mice showed reduced expression of Fst only in early development, suggesting a role for Wnt4 in initiation, but not maintenance, of Fst expression. Adding another layer of complexity to regulation of Foxl2, we found that its action is modulated by SUMOylation, which affects its stability, localization, and activity. We have now examined the effect of over-expression of Foxl2. We created transgenic mice over-expressing Foxl2 under the control of an Amhr2 promoter, active starting early in embryonic development. The transgene up-regulates genes downstream of Foxl2 in the ovary and down-regulates male-determining genes in the testis. A larger follicle pool was also seen in transgenic ovaries. By contrast, the maturation of follicles was normal, with the decrease of follicle numbers during aging proceeding at the same rate as in the wild-type. The over-expression of Foxl2 thus seems to have a primary effect on the initial formation of follicles. This finding is especially relevant to the role of Foxl2 in determining the ovarian reserve and thus, the age of menopause. In the reproductive system, Foxl2 is the only gene expressed uniquely in the ovary. However, Foxl2 is also expressed in the pituitary and in the eye of both sexes, and we have investigated its action in these extra-ovarian tissues as well. In the pituitary, we found that Foxl2 is not required for the specification of gonadotropes, which secrete LH and FSH, the endocrine signals that regulate folliculogenesis in the ovary and spermatogenesis in the testis. By contrast, in Foxl2-ablated mice FSH secretion is dramatically reduced and activin is unable to activate Fshb expression. However, a small number of gonadotropes in the ventromedial pituitary of Foxl2 mutant mice still maintain some FSH expression, suggesting that an auxiliary mechanism independent of FoxL2 and activin can drive some Fshb transcription. Nevertheless, Foxl2 function aeems required in the pituitary for normal expression of FSH.In the eye, we investigated the role of Notch signaling in the formation of corneal and eyelid stroma in mice. We found that over-expression of Notch1 intracellular domain (N1-ICD) impaired eyelid levator smooth muscle formation by down-regulating Foxl2. This is similar to the effect of haploinsufficiency of FOXL2 in humans, which results in type II BPES (blepharophimosis, ptosis and epicanthus inversus syndrome). These data strongly indicate that low levels of Notch1 is crucial for proper Foxl2 expression in periocular mesenchymal cells, which are in turn essential for normal eyelid development. In an independent line of research, we have investigated the role of another forkhead transcription factor, Foxo3, which is not involved in the formation of ovarian follicles but acts to maintain them. Foxo3 is known from work in other systems to act in several biologic processes including cell cycle control, maturation, survival and apoptosis. In the ovary Foxo3 has a specific role in controlling the activation of primordial follicles, governing the size of the ovarian reserve. In its absence, others have shown that the entire cohort of primordial follicles activates and grows uncontrollably, leaving the ovary empty and female mice completely sterile by 15 weeks of age. We have now found that in young females the expression of a mutated form of Foxo3 that cannot be inactivated by phosphorylation causes a delay of follicular growth, with longer survival of primordial follicles, a lower expression of factors involved in follicle maturation, and increased fertility. We also studied adult, premenopausal, and perimenopausal mice, and showed that when active Foxo3 is present, aging processes are slowed and the ovary retains a larger pool of primordial follicles. Furthermore, gene expression levels of markers for further development and aging are consistent with either a developmental delay or slower aging in transgenic ovaries. In addition, we measured the production of gonadotropins, whose levels increase as menopause approaches (an assay method used clinically in humans to diagnose the onset of menopause). We found a higher level of gonadotropins in wild-type compared to age-matched transgenic mice, supporting the interpretation of slower aging in the transgenics. Finally, we studied the fertility of aging mice up to 12 months of age. With aging and approach of menopause, transgenic females showed a greater fertility (31-49%) compared with wild-type animals. This is the first gain-of-function change in a gene that gives greater fertility throughout the reproductive lifespan, and suggests a possible point for pharmaceutical intervention. We also compared the effects of loss of another transcription factor, Lhx8, highly expressed in oocytes. Lhx8 loss causes germ cell loss that is complete by puberty (21 days of age), but the mechanism of death of oocytes and follicles was poorly understood. We found that in Lhx8 mutant ovaries apoptosis does not seem to be involved in the death of oocytes. Germ cells die by activation of autophagy, involving impairment of vascularization and the formation of a nutrient starvation/oxidative stress microenvironment.